license: mit
task_categories:
- tabular-classification
tags:
- customer-retention
- customer-churn
- purchase-prediction
- synthetic-data
- machine-learning
- data-science
- feature-engineering
pretty_name: CRROS Customer Behavior Dataset
size_categories:
- 100K<n<1M
CRROS Customer Behavior Dataset
This dataset is part of my Customer Retention & Revenue Optimization System (CRROS) project. The goal of the project is to simulate realistic customer behavior and use it to build an end-to-end customer analytics workflow, from raw data all the way to business decisions.
Instead of generating completely random records, the dataset follows business-driven rules that simulate how customers interact with products, make purchases, become inactive over time, and eventually churn. This makes it useful for practicing real-world data science workflows while keeping the data completely synthetic.
What's Included?
The repository contains six CSV files that represent different stages of the project.
| File | Description |
|---|---|
customers.csv |
Customer profile and demographic information. |
products.csv |
Product catalog used throughout the simulation. |
transactions.csv |
Purchase history generated from simulated customer behavior. |
interactions.csv |
Customer engagement events such as website visits and marketing interactions. |
customer_features.csv |
Customer-level features created through feature engineering. |
modeling_dataset.csv |
Final dataset prepared for machine learning models. |
Project Objective
I built CRROS to simulate a complete customer analytics pipeline rather than just training a machine learning model.
The project covers:
- Customer behavior simulation
- Data validation and exploration
- SQL-based feature engineering
- Exploratory Data Analysis (EDA)
- Customer churn prediction
- Purchase probability prediction
- Customer targeting and business optimization
- Revenue impact estimation
The idea was to build something that reflects how an end-to-end data science project looks in practice.
How the Dataset Was Created?
The dataset is entirely synthetic, but it wasn't generated randomly. I used NumPy's Random Number Generation tool to design my dataset.
A set of business rules drives customer behavior throughout the simulation. Customers have different value segments, purchasing habits, engagement patterns, and inactivity levels. Those behaviors influence transactions, interactions, and eventually churn.
To make the data more realistic, the simulation also includes:
- Multiple customer behavior patterns
- Behavioral relationships between tables
- Missing values
- Outliers
- Natural variation and noise
This creates a dataset that is much closer to what analysts and data scientists work with in real projects.
Suggested Use Cases
This dataset can be used for a variety of data science and machine learning tasks, including:
- Customer churn prediction
- Purchase prediction
- Customer segmentation
- Feature engineering
- Exploratory Data Analysis (EDA)
- SQL practice
- Machine learning projects
- Business Intelligence dashboards
- Portfolio projects
- Teaching and learning data science concepts
Notes
This is a synthetic dataset created for educational and portfolio purposes. It does not contain any real customer information.
The focus of the project is to demonstrate how realistic business logic can be used to create meaningful datasets for analytics and machine learning workflows.
Resources
If you'd like to see the complete project or learn how the dataset was built, you can explore the following resources:
- GitHub Repository: Customer Retention & Revenue Optimization System
- Project Walkthrough: How to Identify High-Value Customers and Maximize Revenue with Data Science?
Thanks for checking out the dataset! I hope it helps you learn something new or build something interesting. If you use it in one of your own projects, I'd love to see what you create.